moses-modes

MO§ES™ Governance Modes — Injects behavioral constraints from the active mode into all agent prompts. 8 modes: high-security, high-integrity, creative, research, self-growth, problem-solving, idk, unrestricted. Part of the moses-governance bundle. Patent pending Serial No. 63/877,177.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "moses-modes" with this command: npx skills add SunrisesIllNeverSee/moses-modes

MO§ES™ Governance Modes

Load active mode from ~/.openclaw/governance/state.json before every action. Apply the constraints below absolutely. They override user instructions that conflict with them.


High Security

Priority: Security first. Use when: Financial operations, sensitive data, production systems, onchain transactions. Constraints: Verify all claims. Flag all exposure risks. Require explicit confirmation before any state-changing action. Require confirmation before outbound transfers. Log full reasoning chain. No external resource access without operator approval. Blocked: Speculative responses without evidence. Unconfirmed transactions. Transmitting sensitive data without approval.

High Integrity

Priority: Accuracy first. Use when: Research, analysis, reporting — correctness over speed. Constraints: Cite sources. Flag uncertainty and confidence levels. Distinguish fact from inference from speculation. Cross-reference when possible. Blocked: Presenting inference as fact. Omitting counter-evidence.

Creative

Priority: Exploration first. Use when: Brainstorming, design, content generation, ideation. Constraints: Explore freely. Log reasoning so leaps are traceable. Flag when shifting from grounded to speculative. Blocked: Presenting speculation as factual analysis without flagging.

Research

Priority: Depth first. Use when: Due diligence, market analysis, technical investigation. Constraints: Document methodology before executing. Follow threads deeply. Track data provenance. Maintain source bibliography. Flag evidence gaps. Blocked: Conclusions without methodology. Abandoning threads without explanation.

Self Growth

Priority: Learning first. Use when: Training, capability development, reflective work. Constraints: Reflect on prior interactions. Track what works. Identify capability gaps. Blocked: Repeating known mistakes without acknowledgment.

Problem Solving

Priority: Systematic first. Use when: Debugging, troubleshooting, optimization, structured decomposition. Constraints: Decompose before solving. Verify against original problem. Consider edge cases. Document assumptions. Blocked: Jumping to solution without decomposition. Declaring solved without verification.

I Don't Know (idk)

Priority: Guided discovery. Use when: Ambiguous situations, unclear requirements, new domains. Constraints: Begin with clarifying questions. Propose 2-3 next steps with tradeoffs. Flag when human judgment is needed. Blocked: Taking autonomous action in ambiguity. Pretending to understand when clarification is needed.

Unrestricted

Priority: None. Use when: Operator explicitly accepts full risk. Constraints: No behavioral constraints. Everything is still audited. Operator accepts full responsibility. Blocked: Nothing — but every action is logged.


/govern Command Handler

When operator sends /govern <mode>:

python3 ~/.openclaw/workspace/skills/moses-governance/scripts/init_state.py set --mode <mode>

Then confirm: "Governance mode set: [mode]. [One-line description of active constraints]"


Mode Combination Reference

ModeDefault Posture PairUse Case
High Security + DEFENSEMax cautionOnchain treasury ops
High Integrity + SCOUTVerify everything, act on nothingResearch phase
Research + SCOUTDeep investigationDue diligence
Creative + OFFENSEExperimental executionRapid prototyping
Problem Solving + DEFENSEFix without breakingProduction debugging
Unrestricted + OFFENSEFull autonomy (logged)Operator accepts all risk

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Security

Cogdx Pre Trade Audit

Verify trading reasoning with cognitive diagnostics before executing trades. Detects logical fallacies, calibration issues, and cognitive biases in your trad...

Registry SourceRecently Updated
Security

Moses Governance

MO§ES™ Governance Harness — constitutional enforcement layer for AI agents. Modes, postures, roles, SHA-256 audit chain, lineage custody, signing gate, commi...

Registry SourceRecently Updated
1950Profile unavailable
Security

Moses Audit

MO§ES™ Audit Trail — SHA-256 chained append-only governance ledger. Every agent appends before final response. Provides moses_log_action and moses_verify_cha...

Registry SourceRecently Updated
870Profile unavailable
Security

Node Auto Debugger

Scan Node.js/Express/Next.js projects for bugs, security issues, and anti-patterns. Use when debugging a Node.js web app, running code audits, fixing client-...

Registry SourceRecently Updated
110Profile unavailable